Service level agreement based storage access

Abstract

A method and a system implementing a service level agreement based storage access system. A service level agreement based storage access system presents a single interface for data storage consumers and translates generic data operation requests to data operation request specific to a storage server. The SLA based storage access system also monitors storage server performance and may throttle processes to ensure service level agreements are not violated.

What is claimed is: 
     
       1. A method comprising:
 calculating service level agreement performance data, based on a time delay between communicating a request to a storage server and receiving a response from the storage server at a data storage consumer; 
 detecting a service level agreement violation of a service level agreement, based on the service level agreement performance data; and 
 throttling data storage server resources to maintain the service level agreement of the data storage consumer based on the service level agreement performance data indicating the service level agreement violation. 
 
     
     
       2. The method of  claim 1 , further comprising issuing an alert ticket responsive to detecting the service level agreement violation. 
     
     
       3. The method of  claim 2 , wherein the alert ticket includes an indication of the service level agreement violation and the service level agreement performance data. 
     
     
       4. The method of  claim 2 , wherein the issuing the alert ticket further comprises:
 causing display of a violation notification at the data storage consumer, the violation notification including an indication of a time when the service level agreement violation occurred. 
 
     
     
       5. The method of  claim 2 , wherein the alert ticket indicates an amount and a type of resource to throttle to maintain the service level agreement. 
     
     
       6. The method of  claim 5 , wherein the alert ticket includes a suggested resolution to prevent the service level agreement violation. 
     
     
       7. The method of  claim 1 , wherein the throttling the data storage server resources comprises:
 limiting a read per second and a write per second; 
 increasing a read latency and a write latency; and 
 adjusting a level of data reliability. 
 
     
     
       8. A system comprising:
 a service level agreement monitoring module comprising one or more processors to: 
 calculate service level agreement performance data based on a time delay between communicating a request to a storage server and receiving a response from the storage server at a data storage consumer; and 
 detect a service level agreement violation of a service level agreement, based on the service level agreement performance data; 
 a balancing module comprising one or more processors to: 
 throttle data storage server resources to maintain the service level agreement of the data storage consumer based on the service level agreement performance data indicating the service level agreement violation. 
 
     
     
       9. The system of  claim 8 , wherein the service level agreement monitoring module is further configured to issue an alert ticket responsive to detecting the service level agreement violation. 
     
     
       10. The system of  claim 9 , wherein the alert ticket includes an indication of the service level agreement violation and the service level agreement performance data. 
     
     
       11. The system of  claim 9 , wherein the issuing the alert ticket by the service level agreement module further comprises:
 causing display of a violation notification at the data storage consumer, the violation notification including an indication of a time when the service level agreement violation occurred. 
 
     
     
       12. The system of  claim 9 , wherein the alert ticket indicates an amount and a type of resource to throttle to maintain the service level agreement. 
     
     
       13. The system of  claim 8 , wherein the throttling the data storage server resources by the balancing module comprises:
 limiting a read per second and a write per second; 
 increasing a read latency and a write latency; and 
 adjusting a level of data reliability. 
 
     
     
       14. The system of  claim 8 , wherein the one or more processors of the service level agreement monitoring module are further configured to:
 monitor the service level agreement performance data over a predefined period; 
 calculate a usage trend based on the service level agreement performance data over the predefined period; 
 detect a potential violation of the service level agreement based on the usage trend over the predefined period; and 
 issue an alert ticket based on the potential violation. 
 
     
     
       15. The system of  claim 14 , wherein the alert ticket issued by the service level agreement monitoring module indicates the potential violation of the service level agreement. 
     
     
       16. The system of  claim 8 , wherein the throttling the data storage server resources comprises:
 limiting a read per second and a write per second; 
 increasing a read latency and a write latency; and 
 adjusting a level of data reliability. 
 
     
     
       17. A non-transitory machine-readable medium comprising stored instruction, wherein the instructions, when executed, cause a machine to perform operation comprising:
 calculating service level agreement performance data, based on a time delay between communicating a request to a storage server and receiving a response from the storage server at a data storage consumer; 
 detecting a service level agreement violation of a service level agreement, based on the service level agreement performance data; and 
 throttling data storage server resources to maintain the service level agreement of the data storage consumer based on the service level agreement performance data indicating the service level agreement violation. 
 
     
     
       18. The non-transitory machine-readable medium of  claim 17 , wherein the instructions, when executed, cause a machine to perform operations further comprising:
 issuing an alert ticket responsive to detecting the service level agreement violation. 
 
     
     
       19. The non-transitory machine-readable medium of  claim 17 , wherein the instructions, when executed, cause a machine to perform operations further comprising:
 monitoring the service level agreement performance data over a predefined period; 
 calculating a usage trend based on the service level agreement performance data over the predefined period; 
 detecting a potential violation of the service level agreement based on the usage trend over the predefined period; and 
 issuing an alert ticket based on the potential violation. 
 
     
     
       20. The non-transitory machine-readable medium of  claim 17 , wherein the alert ticket indicate the potential violation of the service level agreement.